1. Field of the Invention
This invention relates to an optical wavelength multiplex transmission method which uses a band around a zero dispersion wavelength of an optical fiber and an optical dispersion compensation method for compensating for waveform degradation by a synergetic effect (hereinafter referred to as SPM-GVD effect) of self phase modulation (SPM) and chromatic dispersion (group velocity dispersion: GVD) which is one of several restrictive factors to the transmission distance and the transmission rate in a long-haul, very high-speed optical communication system which employs, for example, an erbium-doped optical fiber amplifier (Erbium-Doped Fiber Amplifier, hereinafter referred to as EDFA).
2. Description of the Related Art
In response to a remarkable increase in the amount of information in recent years, a communication system of a large capacity has become required, and investigations for construction of large capacity communication systems are frequently performed.
For realization of a large capacity communication system, realization by an optical communication system is considered most promising. At present, an optical amplifier multi-repeater system which employs EDFAs is being put into practical use together with, for example, a 2.4 Gb/s optical communication system. In the future, it is forecast that the amount of information will increase progressively as the information-oriented trend advances. It is therefore demanded to build up an optical communication system of an increased capacity corresponding to such increase of the amount of information.
Various methods are available to increase the capacity of an optical communication system, including a TDM (time-division multiplexing) method which involves multiplexing on the time base in order to increase the transmission rate, and a WDM (wavelength-division multiplexing: wherein the wavelength spacing is comparatively great and is called WDM, and wavelength-division multiplexing which involves high concentration multiplexing and is called FDM (frequency-division multiplexing)) method which involves multiplexing on the optical wavelength base.
Of the available methods, a multiplexing method like the TDM method requires an increase of speed of operation of electronic circuits in a transmitter and a receiver in order to increase the transmission rate. At present, several tens Gb/s is considered to be the limit to the speed of operation.
In contrast, with the WDM (FDM) method which makes use of the wide band property of an optical fiber, an increase of capacity to several tens to several hundreds Gb/s is possible by simultaneous exploitation of an increase of the transmission rate, and also the burden to electronic circuits is reduced since multiplexing and demultiplexing are performed simply in an optical region by means of an optical multiplexing apparatus and an optical demultiplexing apparatus (MUX/DEMUX) which employ optical couplers, optical filters and like elements.
In the WDM (FDM) method which involves wavelength multiplexing on the optical frequency base, however, an available band is restricted by gain band dependency of an optical amplifier or wavelength dependency of an optical part. Accordingly, in order to achieve an increase in capacity by multiplexing, the channel spacing must necessarily be decreased to decrease the bandwidth indicated by all channels. Further, in optical transmission of multi-Gigabits, the wavelength of an optical signal must necessarily be set in the proximity of a zero dispersion wavelength of an optical fiber since, otherwise, waveform degradation is caused by chromatic dispersion of the optical fiber.
In an optical communication system to which the WDM (FDM) method is applied in order to achieve such an increase in capacity as described above, however, if the channel spacing is decreased (taking the bandwidth into consideration) and optical signals are set in the proximity of a zero dispersion wavelength of the optical fiber (taking the chromatic dispersion into consideration), an influence of a non-linear effect of the optical fiber, particularly of four wave mixing (hereinafter referred to as FWM), becomes significant, and there is a subject to be solved in that the transmission may be disabled by crosstalk from another channel by such FWM. A similar subject resides in another case wherein wavelength multiplex transmission must be performed in a band in the proximity of the zero dispersion wavelength in order to achieve, for example, upgrading of an existing transmission line.
Meanwhile, as a factor of degradation of the transmission characteristic in the optical amplifier multi-repeater WDM method which particularly makes use of a band in the proximity of a zero dispersion wavelength of an optical fiber, crosstalk by FWM mentioned above is pointed out. The occurrence efficiency of such FWM depends upon the relationship between the zero dispersion wavelength of the optical fiber transmission line and the arrangement of channels.
Three characteristics including: 1. a zero dispersion wavelength, 2. a deviation in zero dispersion wavelength and 3. a dispersion slope (second-order dispersion) are listed as required characteristics for an optical fiber in the WDM method. Those characteristics are closely related to five factors including: a. wavelength multiplexing signal bandwidth, b. gain bandwidth of the EDFA among various optical amplifiers, c. guard band for suppressing FWM (to which the present invention is directed), d. limitation bandwidth by an SPM-GVD effect, and e. presence or absence of an inserted optical dispersion compensator.
By the way, as factors which restrict an increase in distance and an increase in speed of an optical communication system, there are limitations of the loss by an optical fiber and bandwidth limitation by chromatic dispersion. The loss limitation has been almost solved by realization of EDFAs, and it is possible to build up a very long-haul optical communication system for several thousand km or more.
However, the repeater span in a multi-repeater optical amplification system is restricted principally by two factors including: 1. optical SNR (signal to noise ratio) degradation caused by accumulation of ASE (spontaneous emission) from optical amplifier-repeaters, and 2. waveform degradation by an SPM-GVD effect caused by a Kerr effect.
It is already known that, of the two factors, the waveform degradation by an SPD-GVD effect can be compensated for using an optical dispersion compensator having a dispersion value of the opposite positive or negative sign to that of the optical fiber transmission line, and the waveform degradation by an SPM-GVD effect and a dispersion compensation effect can be simulated readily by solving a non-linear Schroedinger equation using the split-step Fourier method.
An optical dispersion compensator used for the object described above is required to cope with a dispersion amount of an optical fiber of a corresponding repeater section and to allow reduction of the number of steps and of the time necessary to realize an optimum dispersion compensation amount and reduction of the cost. Further, the optical dispersion compensation technique is important not only for a 1.55 .mu.m dispersion shifted fiber (hereinafter referred to as DSF) transmission line network being laid at present but also for a long-haul, very high-speed optical communication system and an optical communication system of the WDM (FDM) method which make use of an existing 1.3 .mu.m zero dispersion single mode fiber (hereinafter referred to as SMF) transmission line network.
In a very long-haul optical communication system for several thousand km or more, it is considered desirable to use the zero dispersion wavelength .lambda..sub.0 of the optical fiber transmission line in order to prevent the dispersion penalty and to use the ordinary dispersion region (dispersion value D&lt;0) of the optical fiber in order to minimize the non-linear effect. In order to satisfy the two contradictory requirements, a countermeasure has been proposed which makes use of the ordinary dispersion region for the transmission line and employs an optical dispersion compensator to reduce the apparent dispersion value equal to zero. The optical dispersion compensation technique is effective not only for DSF transmission but also for SMF transmission having a high dispersion value of approximately 18 ps/nm/km.
Various types of optical dispersion compensators have been proposed including dispersion compensating fiber type optical dispersion compensators, transversal filter type optical dispersion compensators and optical resonator type optical dispersion compensators. At present, a dispersion compensating fiber is considered promising from its advantage in that the dispersion compensation amount can be adjusted readily by varying the length of the fiber, and dispersion values higher than -100 ps/(nm.multidot.km) have been obtained by contriving the profile of the core.
The zero dispersion wavelength of an actual optical fiber transmission line presents a deviation in a longitudinal direction. Further, in an optical communication system on land, since it is difficult to set the repeater span to a fixed value (as in a submarine optical communication system), the dispersion amount is not always fixed among different repeater sections. Therefore, ideally an optical dispersion compensator having an optimum dispersion compensation amount is inserted into each repeater section after an actual dispersion amount is measured for the repeater section. However, there is a subject in that such operation requires a great number of steps of operation, long time and a high cost to realize optimum optical dispersion compensators including measurement of dispersion amounts.